Clean up the Perlin noise benchmark #12270
Conversation
|
All the style fixes look great to me, but did you see a perf improvement when siwtching from |
| permutations: permutations, | ||
| } | ||
| let mut rgradients = [Vec2 { x: 0.0, y: 0.0 }, ..256]; | ||
| for i in range(0, 256) { rgradients[i] = random_gradient(&mut rng); } |
There was a problem hiding this comment.
Why not iterators? for x in rgradients.mut_iter() { *x = random_gradient(&mut rng); }
|
@alexcrichton Although this version does run faster than the prior one, it's very slight... only two hundredths of a second faster, on average, on a machine that is pretty slow to begin with. I had no idea that println! was buffered, I do much prefer the look of it, so I'll make that change. |
|
Actually, the switch from |
| @@ -8,128 +8,112 @@ | |||
| // option. This file may not be copied, modified, or distributed | |||
| // except according to those terms. | |||
|
|
|||
| // Perlin noise benchmark from https://gist.github.com/1170424 | |||
| // Multi-language Perlin noise benchmark. | |||
There was a problem hiding this comment.
Check out the link in the next line, there's D, Nimrod, Go, C, C#, lots more. Lots of the structure of this code is the way it is in order to remain easily comparable to these other implementations.
|
r=me after squashing |
Mostly just style fixes, but also remove a heap allocation and switch to using a buffered writer rather than doing 60,000 `println!`s.
`lint_groups_priority`: ignore lints & groups at the same level Fixes rust-lang#12270 changelog: none
Mostly just style fixes, but also remove a heap allocation and switch to using a buffered writer rather than doing 60,000
println!s.